Unmanned Systems Technology 010 | nuTonomy driverless taxi | Embedded computing | HFE International marine powertrain | Space vehicles | Performance monitoring | Commercial UAV Show Asia report

71 Performance monitoring | Focus will increase from the current 80,000 units to about 2.7 million units by 2025, economies of scale will lower costs. Also, if a solid-state sensor (especially a MEMS) is built using a standard or semi-custom silicon chip process, the price of the sensor can drop dramatically once it is manufactured in volumes of hundreds (or even tens) of thousands of units. Using a standard process also has the advantage of being able to integrate more of the signal processing electronics on the chip. However, the design of such an integrated sensor must be accomplished under some restrictions. Unmanned ground vehicles It is not just airborne platforms that are benefiting from improved and more cost-effective sensors. With autonomous ground vehicles such as driverless cars and tractors for example, the number and types of sensors required grows, with vision, radar, and Lidar being chief among these. In automotive batteries, much of the focus has been on monitoring conditions that have an impact on battery lifespan and vehicle range. The key parameter here is determining the battery’s state of health and the state of charge, and sensors play a crucial part in that. There is also an increasing need to monitor internal combustion engines, with more cars being retrofitted to be autonomous (see In Conversation, page 20), and agricultural equipment such as autonomous tractors where the optimum torque and endurance of the engine is needed. Linking the engine sensors to the autonomous control system can identify problems when they occur and, more important, anticipate them. In applications such as agriculture, this allows a tractor to be recalled to base for maintenance before it gets stuck out in a field. One of the most common automotive sensors is the mass airflow (MAF) type, which measures the amount of air an engine takes in. It not only measures the volume of air but also compensates for its density. For research into autonomous vehicles, many manufacturers use engine control systems based around this type of sensor. There are two common designs of MAF sensors used in modern vehicles. One produces a variable voltage output (analogue), the other a frequency output (digital). In both cases their operation is similar, and both types of output can be measured by a voltmeter or frequency counter respectively. Both designs work on the hot wire principle, where a constant voltage is applied to a heated film or wire. This film/wire is positioned in the air stream or an airflow-sampling channel, and is heated by the electrical current that the voltage produces. As air flows across it, it cools down. The wire/film has a positive temperature coefficient resistor, which means the resistance increases with temperature, allowing less current to flow through it to maintain a programmed temperature. The current is turned into a frequency or voltage that is sent to a computer and interpreted as airflow. Adjustments for air temperature and humidity are taken into consideration since they also affect the wire/film’s temperature. To measure engine speed and crankshaft position, a variable reluctance sensor is sometimes used. This is an electromagnetic device designed for use with control units that accept analogue inputs. Traditionally it consists of a permanent magnet surrounded by a winding of wire and, for application flexibility, it can be used in conjunction with a ferrous target that has notches or teeth. Rotation of the target wheel near the sensor’s tip changes the magnetic flux, creating an analogue voltage in the sensor coil that is proportional to rotational speed and position. The voltage is sent to the control unit, where it is often converted to a digital signal to support the controlling function. Position sensors are often based on integrated adaptive threshold Hall effect technology. They sense the speed and position of a target created with various sense features that are machined, stamped, drawn or magnetically encoded. Unmanned Systems Technology | October/November 2016 Monitoring the diesel engine in this autonomous tractor allows it to return to base before any problems occur (Courtesy of New Holland)